Unravelling of the health effects of polyphenols is a complex puzzle complicated by metabolism

• The complexity of the polyphenol mixture in plant foods complicates the search for the most active compounds for health.
• Approaches followed in the last two decades to reduce this complexity were only partly successful.
• Metabolism has to be taken into account and metabolomics will be a valuable tool.

Plant metabolism creates complex mixtures of polyphenols in plant foods. Epidemiology and human trials reduced this complexity, by studying specific foods; subclasses of polyphenols; individual polyphenols, or total antioxidant capacity (TAC). This implies the following assumptions: (1) a limited number of potent polyphenols exists; (2) well-defined natural potent mixtures of polyphenols exist; (3) polyphenols share a common biological activity (e.g. antioxidant activity). To find potent polyphenols (1st assumption), in vitro screening has been widely applied, but most published results are of limited use because metabolism, changing biological activity profoundly, has frequently not been considered. The abundant anecdotal evidence for natural potent mixtures of polyphenols (2nd assumption) on the internet is very hard to verify. Additionally, cross-cultural studies have revealed the potency of e.g. cocoa. Polyphenols share the antioxidant phenolic group which inspired researchers to measure their antioxidant activity, thus greatly reducing complexity (3rd assumption). Unfortunately, the elegant antioxidant hypothesis has to be rejected, because poor absorption and extensive metabolism annihilate any contribution to the endogenous body antioxidants. In conclusion, the above assumptions are hard to verify, and no quick answers are to be expected. Future research should focus on structure–activity relations at nanomolar levels and explore metabolomics.